| Age | Commit message (Collapse) | Author | Lines |
|---|
|
Provide better names for builtin deriving-generated attributes
First attempt at fixing #49967
Not in love with any choices here, don't be shy if you aren't happy with anything :)
I've tested that this produces nicer names in documentation, and that it no longer has issues conflicting with constants with the same name. (I guess we _could_ make a test for that... unsure if that would be valuable)
In all cases I took the names from the methods as declared in the relevant trait.
In some cases I had to prepend the names with _ otherwise there were errors about un-used variables. I'm uneasy with the inconsistency... do they all need to be like that? Is there a way to generate an alternate impl or use a different name (`_`?) in the cases where the arguments are not used?
Lastly the gensym addition to Ident I implemented largely as suggested, but I want to point out it's a little circuitous (at least, as far as I understand it). `cx.ident_of(name)` is just `Ident::from_str`, so we create an Ident then another Ident from it. `Ident::with_empty_ctxt(Symbol::gensym(string))` may or may not be equivalent, I don't know if it's important to intern it _then_ gensym it. It seems like either we could use that, or if we do want a new method to make this convenient, it could be on Ident instead (`from_str_gensymed`?)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Like #43008 (f668999), but _much more aggressive_.
|
|
|
|
|
|
|
|
Most of the Rust community agrees that the vec! macro is clearer when
called using square brackets [] instead of regular brackets (). Most of
these ocurrences are from before macros allowed using different types of
brackets.
There is one left unchanged in a pretty-print test, as the pretty
printer still wants it to have regular brackets.
|
|
|
|
|
|
Implement RFC#1559: allow all literals in attributes
Implemented rust-lang/rfcs#1559, tracked by #34981.
|
|
|
|
|
|
|
|
To allow these braced macro invocation, this PR removes the optional expression from `ast::Block` and instead uses a `StmtKind::Expr` at the end of the statement list.
Currently, braced macro invocations in blocks can expand into statements (and items) except when they are last in a block, in which case they can only expand into expressions.
For example,
```rust
macro_rules! make_stmt {
() => { let x = 0; }
}
fn f() {
make_stmt! {} //< This is OK...
let x = 0; //< ... unless this line is commented out.
}
```
Fixes #34418.
|
|
|
|
|
|
A number of trait methods like PartialEq::eq or Hash::hash don't
actually need a distinct arm for each variant, because the code within
the arm only depends on the number and types of the fields in the
variants. We can easily exploit this fact to create less and better
code for enums with multiple variants that have no fields at all, the
extreme case being C-like enums.
For nickel.rs and its by now infamous 800 variant enum, this reduces
optimized compile times by 25% and non-optimized compile times by 40%.
Also peak memory usage is down by almost 40% (310MB down to 190MB).
To be fair, most other crates don't benefit nearly as much, because
they don't have as huge enums. The crates in the Rust distribution that
I measured saw basically no change in compile times (I only tried
optimized builds) and only 1-2% reduction in peak memory usage.
|
|
derive: Avoid emitting provided PartialEq, PartialOrd methods for c-like enums
derive: Avoid emitting provided PartialEq, PartialOrd method for c-like enums
`ne` is completely symmetrical with the method `eq`, and we can save
rust code size and compilation time here if we only emit one of them
when possible.
One case where it's easy to recognize is when it's a C-like enum. Most
other cases can not omit ne, because any value field may have a custom
PartialEq implementation.
|
|
|
|
When deriving Hash, RustcEncodable and RustcDecodable, the syntax extension
needs a type parameter to use in the inner method. They used to use __H, __S
and __D respectively. If this conflicts with a type parameter already declared
for the item, bad times result (see the test). There is no hygiene for type
parameters, but this commit introduces a better heuristic by concatenating the
names of all extant type parameters (and prepending __H).
|
|
This changes local variable names in all derives to remove leading
double-underscores. As far as I can tell, this doesn't break anything
because there is no user code in these generated functions except for
struct, field and type parameter names, and this doesn't cause shadowing
of those. But I am still a bit nervous.
|
|
This replaces some `if`s with `match`es. This was originally not possible
because using a global path in a match statement caused a "non-constant
path in constant expr" ICE. The issue is long since closed, though you still
hit it (as an error now, not an ICE) if you try to generate match patterns
using pat_lit(expr_path). But it works when constructing the patterns more
carefully.
|
|
Using the same logic as for `PartialEq`, when possible define only
`partial_cmp` and leave `lt, le, gt, ge` to their default
implementations. This works well for c-like enums.
|
|
Also detect unit structs and enums with zero field struct variants.
|
|
`ne` is completely symmetrical with the method `eq`, and we can save
rust code size and compilation time here if we only emit one of them
when possible.
One case where it's easy to recognize is when it's a C-like enum. Most
other cases can not omit ne, because any value field may have a custom
PartialEq implementation.
|
|
|
|
|
|
|
